HYGD: Gold-Standard Annotated Fundus Images for Glaucoma Detection

A collection of fundus images annotated to a gold standard for detecting glaucoma. The dataset is hosted on Kaggle, though specific details on size, collection dates, and authorship are not provided in the available metadata. Its primary purpose is to serve as a benchmark for developing and evaluating automated glaucoma detection algorithms.

Use Cases

• Train image classification models for glaucoma detection based on annotated fundus images.
• Benchmark the performance of different computer vision algorithms against a gold-standard annotated dataset.
• Develop segmentation models to identify specific biomarkers of glaucoma mentioned in the description.
• Validate the generalizability of automated diagnostic tools using a standardized image collection.

Strengths

• Annotations are described as a 'Gold-Standard', suggesting high-quality, reliable labels for model training and evaluation.
• The dataset is specifically designed for a critical medical detection task, indicating focused utility.

Limitations

• Row count and total dataset size are unknown, which may limit suitability assessment for large-scale model training.
• Column-level documentation and sample data are absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality, collection methodology, and potential biases require manual inspection.

Provenance

Source

Kaggle